Multiple audiences (consumers, purchasers, policy makers) want to know about the quality of health care, but developing and reporting quality of care information can be costly. The use of administrative data economizes on data collection costs because the data are usually already available in computer files. Because administrative data are widely available, many organizations use them for performance measurement efforts. However, there has not been a comprehensive effort to characterize the types of quality of care indicators that can be measured with administrative data and how those results may vary systematically from medical record review. The specific aims of the dissertation are the following: (1) Describe the characteristics of indicators that can be assessed appropriately with administrative data; (2) Gauge the accuracy of the administrative data used to construct quality of care indicators relative to the medical record; and (3) Develop a framework to inform decision makers when the additional costs of using medical records are likely to outweigh the benefits of doing so. To accomplish these aims, the candidate will first assess whether or not the data elements required to construct over 600 quality of care indicators can be captured with administrative data. Among a sub-set of those indicators that can be constructed with administrative data, the accuracy of the administrative data will be assessed by comparing the indicator scores based on administrative data to the scores based on the medical record (i.e., gold standard) data. The data for this study come from a large managed care organization located in the Mid-Western region of the country.